Can the Norwegian Oil and Gas Industry Benefit From Communicating their Environmental CSR Efforts?

(1)

MAS5100

Masteroppgave

CSR IN CONTROVERSIAL INDUSTRIES:

Can the Norwegian Oil and Gas Industry Benefit From Communicating their Environmental CSR Efforts?

Høyskolen Kristiania Våren 2020 Innlevering: 23.06.2020

«Denne oppgaven er gjennomført som en del av utdannelsen ved Høyskolen Kristiania.

Høyskolen er ikke ansvarlig for oppgavens metoder, resultater, konklusjoner eller anbefalinger.»

(2)

Preface

This thesis is written as the final part of the MSc in Marketing and Management at Kristiania University College (Oslo). I look back at the past two years with gratitude as it has helped me evolve, expand, and strengthen my knowledge across relevant disciplines. While working on this research study proved challenging at times, it has been an exceptionally educational and rewarding process, leaving me proud of the work I am now handing in.

I especially want to thank my supervisor, Delphine Caruelle, for the invaluable assistance in designing and conducting this study. By her qualified competence in quantitative research, her words of motivation, and her guidance throughout the research process, she has provided me with a good foundation for accomplishing this study. Furthermore, I would like to express my appreciation to those who willingly helped me by completing the survey questionnaire. As well as those who offered their feedback on the questionnaire design, giving me insight and offering alternate explanations of relationships between factors.

June 22. 2020 5051

(3)

Abstract

As corporate social responsibility (CSR) initiatives are used more frequently as a marketing tactic, research is lacking on the subject with regard to controversial industries. The aim of this study is to contribute by adding knowledge about the factors that influence attitude in a setting of environmental CSR messages communicated by the oil and gas industry. The ultimate objective is to examine if and how the Norwegian oil and gas actors should communicate their efforts.

To examine this subject, a 2x2 experimental design was implemented, manipulating four advertisements to demonstrate the conditional factors of message source (internal vs. external) and message framing (positive vs. negative). Attitude was then examined as determined by each experimental condition, accounting for the effects of the mediating and moderating variables.

The study’s findings provide evidence that attitude is influenced by how (message framing) and by who (message source) the message is communicated. However, these relationships are only significant when mediated and moderated by, respectively, the perceived authenticity of the source and by level of environmental concern.

Overall, in order to reach and affect those with higher levels of environmental concern, it is recommended that environmental CSR messages is framed in such a way that it enhances the possible gain of the initiative (positive framing). The message should also be communicated (or supported) by an external third-party, as this enhanced attitude, relative to the internal corporation source. Furthermore, the authenticity of the source proved to be an important mediator and direct predictor for attitude. Focus should therefor lay on presenting CSR efforts in an accurate, truthful, and transparent way.

(4)

1.0 Introduction

This chapter will introduce the topic of the thesis and its relevance. Continuing, it will present the purpose and the positioning of the problem, supplemented by research questions, before the thesis’ outline is presented.

1.1 Introduction and relevance

Never before has the issue of climate change received more attention. The percentage of people concerned about climate change has varied over the years but reached its all-time high in 2019 (Kantar 2020, 8). With 49 percent of the Norwegian population regarding climate change as the most pressing challenge Norway is facing today (Livgard 2019), it becomes evident that the issue is of general concern. Looking closer at differences between age groups, numbers from Kantar indicate that climate engagement is relatively equal across groups, but highest among the younger generation (30 years and younger), with 56 percent ranking climate change as Norway’s most pressing issue. Furthermore, the majority of Norwegians agree that the oil and gas industry should go from producing oil and gas to producing more renewable energy, even if this means lower earnings (65%) and fewer jobs in Norway (57%) (Kantar 2020).

With the increased general attention on climate change, companies across industries have taken steps to reduce their carbon footprint through investments in environmentally conscious activities, or by initiating activities aiming to innovate and improve production processes.

However, far greater challenges await the leadership of corporations producing fossil energy.

Today, 80 percent of the world’s total energy consumption comes from fossil fuels, underlining that the need for innovation and investment in environmentally friendly alternatives is pressing and unceasing (Molstad 2020, 42). In an effort to promote their dedication to this change, some companies in the oil and gas industry has in the latest years gone through name changes to represent the move from oil and gas, and to create associations to energy operations as a whole (e.g., Statoil to Equinor). At the same time, new companies with associations to cleaner energy, rather than oil and gas, have taken over for older companies (e.g., Vår Energi took over operations for ExxonMobile). Whether or not such activities spring from an essential corporate desire to help create solutions for a more environmentally friendly future, or if the steps have been taken to address new social expectations and public concerns about the environment, they can be labeled as corporate social responsibility (CSR) activities, initiatives or efforts.

(9)

CSR received one of the first definitions in 1953 when Bowen defined social responsibilities as “the obligations of businessmen to peruse those policies, to make those decisions, or to follow those lines of action which are desirable in terms of the objectives and values of our society” (1953, in Agudelo, Johannsdottir and Davidsdottir 2019, 4). This definition and the amplified focus on social responsibility raised interest among scholars, and the concept of CSR was approached in various ways in the following decades. CSR has been viewed as the corporate concern about compliance with laws and regulations, as well as activities to secure a healthy working environment, the education, the happiness of employees, and the social welfare of the community (Agudelo et al. 2019, 5). However, in more recent years, the social expectations towards corporations have shifted, where the environment and sustainability have come to dominate both academic literature and corporate activities with regards to CSR efforts (Ghobadian, Money and Hillenbrand 2015, 277; Agudelo et al. 2019, 2).

The reason why corporations engage in CSR activities is not solely to support a good cause. In a study by Yoon, Güran-Canli, and Schwarz, CSR is referred to as activities implemented in order to address consumers’ social concerns, create a favorable corporate image, and develop a positive relationship with consumers and other stakeholders (Yoon et al. 2006, 377). Thereby, you could call CSR a form of regulatory risk management, rather than efforts initiated only for a good cause. Furthermore, O’Connor and Gronewold (2012, 71) identify CSR initiatives in the oil and gas industry to be serving a dual function. CSR is used both to enhance corporate advantage relative to corporations within the same industry and also as “protection from activist groups and industry turbulence” (in Miller and Lellis 2015, 213). Thereby, CSR is activities and efforts initiated to address social concerns with the ultimate goal of protecting the corporation from external turbulence and securing public support. It is these views I will be putting forth in the thesis.

The previous head of the Norwegian oil fund, Yngve Slyngstad, stated that the businesses of tomorrow will be evaluated on more than traditional figures (Molstad 2020, 43). By means of accounting non-financial metrics, corporations will be expected to deliver not only on revenue but also on environmental capital values (Ihlen 2007, 44; Molstad 2020, 43). This statement seems to reflect the expectations of the Norwegian public, where 78% of the population expects the business world to take an active part in limiting the emission of greenhouse gases (Livgard

(10)

and environmental actions (Schultz 2013, 363). This means that organizations should not only work to fulfill owners’ interests and to increase profits, but also provide for the

“commonwealth” in form of societal and environmental contributions.

1.2 Purpose and positioning of the problem

With the oil and gas industry aiming towards the production of cleaner energy and creating environmentally friendly solutions, it would be beneficial for corporations to effectively communicate these efforts in order to improve or maintain consumer attitude towards the firm.

Earlier this year, some of Norway’s most prominent leaders from business, research, politics, and government, gathered at the Sola meeting (Solamøtet). The Sola meeting is an annual business policy conference, where the goal this year was to show the steps that should be taken in order to help solve the global challenges we’re facing today. Present at the meeting was some of the biggest oil and energy actors in Norway (Equinor, Shell, Aibel, Aker Solutions), presenting their solution for a better, more environmentally friendly, future (NHO 2020).

Among the ‘solutions’, Aker Solutions presented a future with floating, offshore wind energy, while Equinor launched an extensive plan to reduce their climate emissions by electrification of the substantial Johan Sverdrup shelf, and presented their aim to achieve an emission-free production, with a 40 percent cut by 2030 (Molstad 2020, 42). However, their pursuit of innovation and the goals these companies are aiming at achieving are not widely broadcasted, at least not to the lay public. Thereby one can wonder why certain companies are reluctant to communicate their CSR initiatives and activities, and I propose the following research problem:

Can controversial industry actors, such as the oil and gas industry, benefit from communicating their CSR efforts and how can these efforts be communicated effectively?

As CSR is an integral part of the external and internal corporate communications (public relations, marketing, advertising, etc.) across industries (Schultz 2013, 363), one would think that communicating CSR could help improve company image. However, this is not necessarily the case as literature on corporate reputation suggests that CSR efforts can be viewed as a form of manipulation and misrepresentation by some (Ihlen 2007, 46). As the oil and energy industry has a controversial image, it can be deemed inevitable that tomorrow’s leaders will encounter problems trying to communicate their efforts, and many already have. As BP in 2019 launched

(11)

a campaign to promote investments in wind energy, they were faced with accusations of greenwashing (presenting environmental claims that are inaccurate). The message was put forth in such a way that the audience could misjudge the size of the actual efforts (Chapman 2019), indicating a lack of knowledge about how to communicate CSR efforts. In another setting, Equinor partnered up with the popular podcast, Forklart, a podcast delivered by one of Norway’s biggest newspapers. This created controversy, as the podcast by some was deemed as stealth marketing (invisible advertising) (Jerijervi 2019).

Gosselt, Rompay and Haske (2019, 422) found that even if CSR motives had been inconsistent, consumers still had a tendency to believe that the corporation had sincere motives when supporting environmental issues, enhancing public attitude towards the effort. However, there is a lack of literature studying the effect of CSR activities on consumer attitude with relations to controversial industries. So, in an effort to narrow the focus onto feasible research questions, I propose that the oil and gas industry faces two issues with regards to environmental CSR communication; who should communicate the CSR efforts and how should they be communicated? Firstly, it would be of interest to examine if the public’s attitude towards the CSR message and the actor presenting the environmental initiative would vary based on who the source of the message is. In fact, empirical findings suggest that communication through an external third-party can enhance the creation of positive attitudes, relative to using an internal corporate source (Groza et al. 2011; Miller and Lellis 2016). Then there is the question of how to best communicate the effort. Just as attitude towards an issue, a corporation, or even a public persona, vary between individuals, the effort it takes to adjust attitude may vary. Research finds that information (e.g. environmental CSR messages) will be perceived differently simply by the way it is expressed and presented (Martin and Marshall 1999, 206). One way to differentiate between ways of conveying messages is through message framing. The theory of message framing states that negatively and positively framed messages may be received differently, even when presenting logically equivalent information (Jones, Sinclair and Courneya 2003, 181).

Furthermore, there have to be some factors regulating the effect of who and how on attitude.

So, given the lack of relevant academic literature on this specific subject, it can be difficult to assess if general CSR literature is applicable to controversial industries. Furthermore, as the oil and gas industry has received a lot of critique through the years, it would be interesting to

(12)

all of the above will be elaborated on in the theory section, we know that any communication has a certain message source and a way of being framed, and we know from prior research that both have the potential to affect attitude formation. Therefore, I propose the following research questions:

RQ1: To what extent can the message source and message framing of a CSR campaign for the oil and gas industry estimate consumer attitude towards the advertisement and the organization running the ad?

RQ2: What mediates and moderates these effects?

1.3 Outline

This thesis is structured five parts; the introduction, the framework, the methodology, and research design, the analysis and results, and discussion and implication. In part one I have introduced the subject of interest and presented the purpose and positioning of the study. The next part will present the framework of the thesis, outlining the theory, empirical research, and perspective on which this thesis is based. By means of theory and empirical findings hypotheses are formed and summarized in the hypothesized model. Following, part three will present the applied methodology and research design. This will give the reader insight into the process of forming and implementing the experimental study design. Part four will present the analysis and results of the study, where the first section of the chapter presents the preparation of data material and initial analyzes of data appropriateness. Moving on, the proposed hypotheses are tested by means of relevant analyzes, creating a picture of how the implemented factors relate to each other. Concluding with discussion and implications, part five presents the main findings, theoretical contributions, and practical implications of the study, while also presenting its limitations and possible avenues for further research.

(13)

2.0 Framework

Building on theories and research that are relevant for the positioning and the problem of the thesis, this chapter will present the theory that forms the conceptual framework. Starting off, the chapter will present the relevancy of two specific theories; The Persuasion Knowledge Model (PKM) and Attribution Theory. These theories have been used in previous research and proven helpful in understanding attitude formation in a setting of CSR messages and efforts.

Furthermore, the chapter presents variables plausible to affect attitude, accompanied by possible moderators and mediators. From these variables, reasonable hypotheses have been formed and then summarized in the hypothesized model.

2.1 Relevant Theories

The concept of corporate social responsibility (CSR) has grown to receive a lot of attention, which has cumulated in several studies aiming to understand the effect CSR efforts have on audiences’ responses and attitudes (e.g. Gosselt et al. 2019). When corporations take the step to communicate their CSR efforts, it will in most cases be a persuasive attempt to produce favorable consumer perceptions. This ultimately involves communicating their actions in an effort to improve the corporate image and create positive attitudes. (Groza, Pronschinske and Walker 2011, 640). So, what is attitude? When using the term attitude, I point to the general evaluation an individual makes about other people, objects, issues (Merriam WebsterA s.v.), or in this case, about a CSR message or a corporation. Prior research states that attitudes can be altered or conditioned via media exposure (Caroll 2013, 125), however, available literature presents varying results in regard to attitude formation in a setting of CSR messages and green advertising.

In a study looking closer at how green advertisements were perceived, results indicated that corporations that actively invested in green solutions were better off not promoting them. That is, a no-advertising strategy resulted in a more positive brand attitude, compared to when the efforts were advertised (Nyilasy, Gangadharbatka and Paladino 2014, 700). In a different study, Gosselt et al. (2019) report that CSR labeling on product packaging indirectly influence attitude, where positive external third-party labels resulted in more positive brand attitudes. With regard to attitude formation, the attributed motives consumers assigned the CSR effort served as a

(14)

studies have found that consumers identified a distinction between self-centered motives (strategic and egoistic motives) and other-centered motives (values-driven or stakeholder- driven). Whatever motives the consumer identified and attributed to the CSR initiative, it had an effect on attitude (Ellen et al. 2006, 154; Miller and Lellis 2015; 75).

However, none of the above-mentioned studies looked into attitude with regards to controversial industries, rather the focus was aimed at companies selling products which actually were, or was claimed to be, more environmentally friendly. As available literature presents varying results in regard to attitude formation in a setting of CSR messages and green advertising, it is important to put forth relevant theories that have been proven useful in explaining previous results. Common to several studies is that scholars have applied the Persuasion Knowledge Model (PKM) and Attribution Theory in an effort to explain their results. These two theories will therefore be utilized in this thesis.

2.1.1 The Persuasion Knowledge Model

A prerequisite to study the formation of attitude towards communication is the underlying assumption that people will, in fact, evaluate persuasion attempts (Friestad and Wright 1994, 16). A number of prior studies have applied the Persuasion Knowledge Model (PKM) in an effort to explain how consumer perceptions of corporate intentions and authenticity affect audience response and attitude (Artz and Tybout 1999;Groza et al. 2011;Gaither and Sinclair 2018). The PKM was first presented in 1994 by Friestad and Wright and is a model aimed at explaining how people’s persuasion knowledge is used to interpret, evaluate and respond to persuasion attempts from advertisers, salespeople, or in this case a corporation (Friestad and Wright 1994, 1).

The model assumes that individuals will access existing knowledge about persuasion, fully or partially, whenever they try to comprehend a message (Friestad and Wright 1994, 2). In doing so they do not only evaluate the message claims, they also use their persuasion knowledge to judge the agent’s (message source) motives and evaluate if the agent is trustworthy and transparent (Friestad and Wright 1994, 3). That is, consumers will elaborate on messages and the source of the message in an effort to develop valid and accurate assumptions about them, and based on these assumptions, consumer attitudes will form (Artz and Tybout 1999, 59).

(15)

The concept of accurate or valid attitudes is central in the works of Friestad and Wright. The model suggests that when individuals are exposed to persuasion attempts (advertisements and corporate communication), they will learn over time what constitutes a persuasion attempt. This experience with and knowledge about the marketplace is stored in the mind of the consumer and can be used as a tool to cope or process future persuasive messages (Evans and Park 2015, 157). By using knowledge cumulated over time through repeated exposures, the consumer has adequate information to form attitudes that are justified and true for them, also referred to as valid and accurate attitudes.

2.1.2 Attribution Theory

Other studies also aiming at explaining consumer responses to CSR initiatives have applied Attribution Theory, often as a supplement to PKM (e.g., Ellen, Webb and Mohr 2006; Miller and Lellis 2015). The theory was first developed in general psychology literature, made to investigate the underlying causal explanations people make when confronted with other people’s social behavior (Wiener 1972, 203). Or put more simply, individuals will try to understand why people do what they do. Research on consumer behavior has implemented attribution theory, suggesting that consumers will engage in similar attributional processing when evaluating corporate, as opposed to individual, behavior (e.g. Nyilasy et al. 2014, 696;

Gosselt et al. 2017, 415).

Given that the behavior is perceived as intentional, the consumer will attribute causes to behavior (Pashler 2013, 84). Thereby, attributional processes can alter attitude formation processes, where consumers that attribute insincere motives to the corporation are less likely to express positive attitudes towards the corporation, vice versa for sincere motives (Nyilasy et al.

2014, 696). In the introduction, there was given examples of oil and energy marketing gone wrong, among these were the case of Equinor’s sponsorship of Aftenposten’s Forklart podcast.

If Attribution Theory was to be applied to this situation one could say that the audience perceived the sponsorship to be strategical (they deemed it to be stealth marketing) and thereby attributing insincere motives to Equinor.

Attribution theory is related to PKM and can in this setting be used to understand how consumers attribute motives to an organization as a means of evaluating a message (Miller and

(16)

is relevant as it affects attitude formation (Friestad and Wright 1994, 16). As CSR efforts easily can be seen as communicative attempts to maintain or improve the public’s attitudes or acceptance of an advertisement, corporation or industry, it becomes important to understand the factors affecting attribution formation and perceptions. Therefore, to further examine audience response to CSR efforts it could be beneficial to apply these two theoretical constructs.

2.2 Message Source

The consumer will often use their preconceived image of and beliefs about a corporation as a guide for their response to corporate actions and communications (Bostdorff and Vibbert 1994, 146). Consequently, the sender of a communicative effort might by itself affect attitudes.

Accordingly, previous literature has shown that consumer evaluation of environmental claims will vary depending on the source making the claim. According to the PKM, individuals will consciously or unconsciously, seek to identify the “perceived persuasion agent”, where the persuasion agent is whoever seems responsible for orchestrating the persuasion attempt (Friestad and Wright 1994, 8). Groza et al. (2011) point to the fact that little research has been done to unveil the effects of information source on attitude formation. In their study, they seek to gain understanding about whether internal firm messages (e.g., annual report or corporate ad) or external third-party messages (e.g., from a news agency) resonate more or less desirable with consumers. The results indicated that individuals did in fact assign attributions to the CSR initiatives differently depending on the message source. Even though the message source did not directly influence attitude formation, the effect was mediated through the motives they attributed to the source (Groza et al. 2011, 645; 648). This indicates that the perceived legitimacy of the organization and the motives assigned to the relevant message might be influenced on the basis of the “source” by which consumers receive the message (Groza et al.

2011, 645). Overall, findings indicated that with proper use of message source, corporations could to some degree manipulate the attribution process, making communications of CSR efforts a viable way of attaining positive corporate associations (Groza et al. 2011, 639).

Similarly, Miller and Lellis (2016) found that the identity used to present energy-focused messages has a significant effect on public perception and response. Results from the study uncovered that the participants assigned attributions to the underlying motivations of the messages based on the source presenting the information. When using the name of the corporation in CSR messages (corporate identity), the public perceived the purpose of the

(17)

message to be strategically driven. That is, as a form of crisis management where the corporation’s intent is to set the status quo and promote their innovations. Conversely, if the message was presented by the corporation, but by using the identity of trade groups that represent the industry sector (industry identity), the public see the message as a form of political advertising and/or as a tool for public education (Miller and Lellis 2016, 84). The study revealed no result as to which kind of message source that most effectively stimulates positive attitudes but gave insight as to what source that effectively conveys the message based on the issue that is to be communicated. If the goal is to generate support for a specific industry issue (in this case, a pipeline carrying oil from oil sands in the U.S.), it will be favorable to use an industry trade group. On the other side, if the goal is to promote the overall benefits of an industry sector to the society (generation if revenue, creation of jobs), it would give a slight advantage to use a corporate identity (Miller and Lellis 2016, 82).

Furthermore, Gosselt et al. (2019) examined consumer attitudes towards brands, perceived corporate credibility, and perceived motives as dependent on the CSR message and the source of behind the CSR claim. Results suggested that attitude is partially dependent on the source of the claim. If the claim made was communicated through an external third-party source, the message proved more effective in producing positive consumer evaluations than claims made through internal non-expert sources (Gosselt et al. 2019, 414). By means of attribution theory, this indicates that consumers attributed more sincere and credible motives to the external source, whereas the internal CSR claim only proved effective if backed up and verified by an external CSR label (Gosselt et al. 2019, 421). With this view in mind, it can be thought that respondents will react differently to advertisements given the message source. Meaning the message source in the ad can be perceived to be accurate and sincere, or they can perceive the given source to be a puppet for another company.

From these studies, it becomes evident that the attributes consumers assign the message source can influence their perception of the source and the message. This effect can to some extent be explained by Friestad and Wrights Persuation Knowledge Model (PKM) (Groza et al. 2011, 641). As mentioned earlier, the model maintains that the target (consumer) will work to cope with a persuasion attempt by seeking to form valid (accurate) attitudes towards the agent (corporation behind the message) based on who is responsible for the message (Friestad and

(18)

in the process of forming valid attitudes towards an advertiser, the consumer will not only form an attitude towards the message but also assign different motives to message source that ultimately will influence the attitude formation. Furthermore, previous research found that the motives an individual attribute to the CSR message is partly dependent on the source of the message (Groza et al. 2011; Miller and Lellis 2016). More specifically, it proved that the use of external third-party source is more effective in generating positive consumer attitudes (Gosselt et al. 2019). Building on this knowledge, I propose the following:

H1: External third-party CSR messages has a more positive impact on consumer attitude than internal CSR messages

2.2.1 Perceived Authenticity

As past research suggests that consumers will attribute motives to CSR messages and the source of the message, it would be beneficial for this study to look at the attributions that Norwegian audiences make. Past research applying the Persuasion Knowledge Model (PKM) to CSR efforts proposes that the public response to CSR messages are predicted, to some extent, by perceptions of source intentions and trustworthiness (Miller and Lellis 2015, 70; Gilbert and Malone 1995, 21). A similar construct to trustworthiness is authenticity, which is “being actually and exactly what is claimed” (Merriam WebsterB s.v.). This definition implies being fully trustworthy and according to facts. Prior research states that organizations that succeed in the creation of an authentic image will in many ways be regarded by the public to be credible, trustworthy, genuine, and honest (Molleda and Jain 2013, 436).

Past research has found that it could advantageous to disclose corporation values, motives, and beliefs in a manner that helps the public assess the identity and integrity of the organization’s actions (Molleda and Jain 2013, 437). It can be argued that organizations have the possibility to enhance their image of authenticity by sharing accurate information with their consumers and other stakeholders. Public announcement of organizational values may encourage the audience to evaluate the organization in accordance to how well it upholds the declared values (Bostdorff and Vibbert 1994, 154). Where higher perceived authenticity helps the corporation

(19)

gain support for its CSR initiatives and the firm in general (Beckman, Colwell and Cunningham 2009, 203). Consequently, it is plausible that audience evaluation of the source’s authenticity (being and doing what is claimed) can mediate attitude formation and change, and that by sharing accurate information, corporations have the opportunity to enhance their image of perceived authenticity (e.g. Molleda and Jain 2013; Nyilasy et al. 2014; Li et al. 2017).

Furthermore, a corporation cannot only rely on sharing accurate information, one also has to ask what kind of CSR efforts that will enhance authenticity. In similar ways as trust and credibility, authenticity is an experience and perception that is co-created by the organization and its stakeholders. It is an ongoing negotiation of meaning and understanding, and therefore in some sense, a function of perceived genuineness that could determine the quality of an organization’s public relations (Mollenda and Jain 2013, 437). In a study published by Li, He, Liu, and Su (2017), they measure consumer perceptions of environmental legitimacy, and much like authenticity (being what is claimed), legitimacy is defined as the degree to which one is exactly as presented (Merriam Websterc s.v.). The purpose of the study was to look closer at how consumers react to organization’s environmental actions, be it easy-to-be-observed (symbolic) actions to secure social support or the adaptation of new practices to improve environmental performance (substantive actions). Through their study, they found that substantive, hard to reverse commitments induced significantly higher perceptions of environmental legitimacy than symbolic actions (Li et al. 2017, 598).

However, another study found that substantive action may be less visible for consumers, thereby making them rely more on symbolic action for information about the corporation (Schons and Steinmeier 2016, 360). The study advocate that with consumers’ limited opportunity to distinguish between symbolic and substantive CSR actions, corporate transparency is essential (Schons and Steinmeier 2016, 366). As the oil and gas industry undoubtedly require a green shift, both for the sake of the environment and because of public demand (Kantar 2020), I propose that symbolic actions alone are insufficient to increase authenticity and influence attitude. When initiating CSR efforts without a true commitment to improving corporate operations, it can lead consumers to perceive initiatives as inauthentic (Beckman et al. 2009, 204), therefore symbolic CSR actions need to be supported by substantive ones (Schons and Steinmeier 2016, 361). This finding can be explained by the

(20)

corporation will be able to communicate their efforts in a way that substantiates their stated values and proving them in their deeds, and thus coming across as more authentic (Molleda and Jain 2013, 436).

As substantive CSR efforts are shown to induce higher perceptions of authenticity (Li et al.

2014) and because there is a high demand for substantial change in the oil and gas industry (Ihlen 2007; Molstad 2020; Kantar 2020), I will for the sake of this study focus the communication to present substantive CSR efforts. Furthermore, building on the fact that high perceived authenticity has proven important in obtaining beneficial corporate outcomes (Molleda and Jain 2013; Miller and Lellis 2015; Gilbert and Malone 1995), I suggest that the attributes consumers assign the message source will mediate the effect CSR communication has on attitude. Thereby I propose the following:

H2: The effect of external CSR messages on consumer attitude is mediated by the perceived authenticity of the source

2.3 Message Framing

People hold attitudes for many different reasons, causing a great variation in what kind of information is needed to affect and potentially alter these attitudes (Petty and Cacioppo 1986, 127). Literature suggests that in order to effectively influence attitudes, corporate CSR messages should make sure to inform the public about the activities the company engages in (Bostdorff and Vibbert 1994, 146; Schultz 2013, 363; Molstad 2020). In some ways, the corporate activities function as a reflection of the corporate values, thereby shaping public perceptions. Some companies convey their CSR activities in a way that brings forth their organizational values and goals, while others use a more fact-based approach to inform the public of their specific efforts to make a difference. Both directions are aimed at forming and affecting the image of the organization, and ultimately the publics’ attitude towards the organization. Whatever information the corporation chooses to convey, the message can be framed in different ways.

(21)

Message framing can be conceptualized as truly equivalent information perceived differently by consumers based upon how it is presented (Martin and Marshall 1999, 206), and can thereby be defined as the way in which a corporation expresses or outlines their message. A framework frequently used to explain message framing is prospect theory. Drawing upon the work of Kahneman and Tversky (1977), prospect theory holds that individuals will respond differently to messages depending on how the message is framed (Jones, Sinclair and Courneya 2003, 181). This implies that messages can be framed either to stress the benefits or potential gain of an initiative (positive/gain message framing), or it can emphasize how the initiative prevents a possible consequence or loss (negative/loss message framing) (Schiffman, Kanuk and Hansen 2012, 281; Jones et al. 2003, 181). With regards to CSR messages, a positive message frame might highlight how an initiative has a beneficial impact on the environment, whereas a negatively framed message might focus on how the corporation’s action will lead to the removal or reduction of something that is a destruction of the environment.

Prospect theory suggests that when considering messages, people opt to favor the negatively framed message that emphasizes the avoidance of a possible loss, rather than the positive frame emphasizing a gain (Chang 2007, 145). A possible reason for this is that people rarely weigh gains and losses uniformly, even if the two are logically equal. Consequently, people will respond differently to factually equivalent messages because of how the message is worded.

Furthermore, people typically are more sensitive to losses than to the counterpart gains, thereby having a general preference of avoiding losses rather than creating gains (O’Keefe 2012, 5).

Thus, message framing is anticipated to have a significant influence on how people will perceive a message and the way in which alternatives will be evaluated (Davis 1995, 286).

Based upon research that suggest that people will avoid loss, rather than achieve a possible gain (Jones et al. 2003; Chang 2007; O’Keefe 2012) I propose the following hypothesis:

H3: Negatively framed CSR messages have a more positive impact on consumer attitude than positively framed CSR messages

(22)

2.3.1 Level of Environmental Concern

Even though negative framing seems to be the preferred, other research has shown that the appropriate framing of a message is not only dependent on the issue that is to be communicated but also target characteristics and attitudes (Schiffman et al. 2012, 281). General research on framing has shown that negative framing tends to be more effective in persuading consumers, than positive framing, when the goal of the message is to influence attitudes and when consumer issue involvement is high (Olsen, Slotegraaf and Chandukala 2014, 123). Individuals that feel highly involved in the issue of the message are more likely to processes the issue-relevant message in detail, thereby being easier to persuade. In such situations, studies have shown negatively framed messages (what is prevented) to be more effective than positively framed messages (what they have achieved) (Grau and Folse 2007, 26). One the other hand, if the individual is less involved with the CSR effort that is being communicated, the message will be communicated more effectively by the use of positive message framing (Grau and Folse 2007, 29). Specifically, a study by Shiv, Britton and Payne (2004,207) found that high issue involvement induces motivation of elaboration. With the motivation to elaborate and understand a message, the negatively framed messages were proven more effective.

It is therefore likely that people highly involved with the specific issue of a message to elaborate on the informational details in the message (Grau and Folse 2007, 26). The reason for this can stem from the fact that involvement, together with perceived personal relevance or responsibility, function as a drive for motivation, making people motivated to understand, learn or evaluate a message (Van Riel and Fombrun 2007, 53; Petty and Cacioppo 1986, 185; Friestad and Wright 1994, 17). Furthermore, the persuasion knowledge model suggests that attitude could be partly related to the inclination people have to interpret persuasion episodes on the basis of personal attitudes towards the topic of the message or the message sender (Friestad and Wright 1994, 23). Additionally, having knowledge about the issues advocated in a message will drive the ability to comprehend the arguments of the message and at the same time providing the individual information to critically evaluate the arguments (Fennis and Stroebe 2016, 194).

Personal relevance, motivation, and knowledge are therefore ever evolving. In relation to CSR efforts and the content of the message (environmental actions), one factor that might affect individuals’ motivation to seek accurate agent attitudes is how relevant or salient the agent is in the mind of the individual (Friestad and Wright 1994, 9). With regard to environmentally conscious people, it is plausible to believe that oil and gas corporations appear more prominent

(23)

in the mind of the individual and that messages from such controversial corporations’ sparks interest. Thereby, it can be assumed that people with high levels of environmental concern will have higher motivation and interest for assessing both the message and the source of the message in order to form attitudes about the ad and the source of the ad.

Bailey, Mishra, and Tiamiyu (2016) attempted to develop and validate a scale to measure consumer receptivity to green messages. They found that the level of receptivity to green ads functioned as a moderator for attitudes, where highly receptive consumers had different attitudes and intentions toward a company and its green initiative than low green receptives.

Furthermore, people receptive to green messaged proved to also be concerned about the environment (Bailey et al. 2016, 339), making a connection between attitude towards campaign and corporation, and the level of environmental concern. One study has found that people highly receptive to green advertising might find companies more trustworthy and respond more favorably to green claims, than low receptives (Bailey et al. 2016, 339). Even so, other research suggests the more environmentally concerned an individual is, the more skepticism they will assert towards green claims (Do Paco, Finisterra and Reis 2012, 153).

Even though there is a lack of research on message framing connected to CSR messages, environmental messages and/or corporate statements, it is prevalent that message framing can cause an impact on communication effectiveness. Research on message framing suggests that it is possible to generate positive consumer attitudes by the use of “correct” message framing (e.g. Grau and Folse 2007; Olsen 2014). Linking personal relevance and involvement to message framing, Martin and Marshall (1997, 212) found that for individuals with low involvement, positive message framing resulted in more favorable attitudes. In contrast, negative message framing was more effective in inducing positive attitudes among highly involved consumers. These results indicate that the felt personal relevance moderates the impact of message framing, where the level of involvement with the issue at hand is of importance when assessing framing effects (Martin and Marshall 1997, 213). Furthermore, previous research examining how individual differences affect attitude towards messages suggests that these individual differences can enhance, reduce, and even eliminate framing effects (Chang 2007, 165). Therefore, when forming and framing a CSR message the corporation should think about who they desire to reach.

(24)

We already know that Norwegians are concerned about the environment, as well as skeptical towards the oil and gas industry (Kantar 2020; Livgard 2019), it is thereby plausible to believe that their personal relevance and level of environmental concern is higher. Additionally, I suggest that environmentally conscious people would be more inclined to assess the CSR message and the source of the message in order to form attitudes (Grau and Folse 2007, 26).

Linking this finding to prior research on message framing, it suggests that negatively framed messages will work more effectively to generate positive attitudes in situations where individuals have high processing motivation (e.g. Martin and Marshall 1997; Shiv et al. 2004;

Shiffman, Kanuk and Hansen 2012; Olsen et al. 2014). Thereby causing the level of environmental concern to moderate the relationship between message framing and attitude, and I propose the following:

H4: The effect of negatively framed CSR messages on consumer attitude is moderated by individuals’ level of environmental concern

2.4 The hypothesized model

The hypotheses to be tested in the present study were constructed based on the research question presented in the introduction and with theoretical findings from the conceptual framework in mind. Together they form the hypothesized model presented in figure 1. The model consists of two independent variables (message source and message framing), while the dependent variable of the model is attitude. Attitude has been divided into two measurable units; attitude towards the advertisement and attitude towards the source of the message. The model assumes that (H1) the use of an external third-party message source will have a positive effect on consumer attitudes. However, this effect is expected to be (H2) mediated by the perceived authenticity of the source. Furthermore, it proposes that (H3) a negative message frame will have a more positive impact on attitude, than that of a positively framed message. This effect is expected to be (H4) moderated by individual level of environmental concern.

(25)

Figure 3.1 Hypothesized Model

Level of

Environmental Concern Framing

condition Source condition

Attitude towards advertisement

Attitude towards source

Attitude

H3 H1 H2

H4 Perceived Authenticity

(26)

3.0 Methodology and research design

This chapter will describe and defend the methodical approach of the study. Firstly, the choice of method and research design is described, before the preliminary study is presented. The preliminary study emphasizes the choice of stimuli and construction of test materials, the distribution of materials, and presents the outcome of the manipulation check. Furthermore, this chapter will provide insight into the design of the main experimental study, where the creation and implementation of the study is presented.

3.1 Research Design

The research questions applied in this study seeks to examine the Norwegian publics’ attitude towards CSR campaigns and towards the industry actor disbursing the advertisements. Attitude is to be measured based on source and framing conditions, accompanied by a possible moderator and mediator. The study proposes that negative message framing and the use of an external third-party source will result in more positive consumer attitudes. Furthermore, aiming to understand the factors affecting attitude, hypotheses suggest that these effects will be mediated and moderated, respectively, by the perceived authenticity of the corporation and the individual level of environmental concern.

To allow for measurement of various views and opinions on CSR campaigns in a controversial industry, a quantitative approach is chosen (Malhotra, Birks and Wills 2012, 187). Furthermore, an experimental causal research design is applied to investigate the cause-and-effect relationships within the data (Malhotra et al. 2012, 371). By the use of causal experimental design, the thesis applies manipulation of the independent variables (message source and message framing) in order to answer the relevant hypotheses.

(27)

3.2 Preliminary Study - Pretest of Manipulation Fit

To ensure that the CSR messages would appear different in the eyes of respondents, there was conducted a pretest to control for the appropriateness of manipulation stimuli. The respondents were presented with factually equivalent messages, apart from slight changes in wording to reflect message framing manipulations and the use of different company logos to reflect message source. By assessing the effectiveness of manipulation, one allows for greater certainty that manipulations are suitable for further use in the main study (Ejelov and Luke 2019, 7).

3.2.1 Stimuli development

To manipulate both message frame and the source of the message, four separate fictitious advertisements where constructed. Based on previous studies applying framing and source manipulations (examples of message framing in appendix A) different versions were created and presented to an expert source before arriving at the final four advertisements (figure 4.1).

For the manipulation of message framing, advertisements were composed either to stress the beneficial impact of the CSR effort (e.g. more renewable energy) or how the CSR effort leads to avoidance of potential harm to the environment (e.g. less non-renewable energy). It is crucial that the manipulated CSR communication contain evident actions and clear statements of the potential outcome, while the information presented is equal in impact (Davis 1995, 286). When designing environmental CSR communication, the message (either negatively or positively framed) will ultimately present an initiative that is beneficial for then environment. This posed a challenge when wording the messages as most literature applying message framing is directed towards the individual actions of the consumer. Where positive framing stresses the potential gain of taking action and negatively framed massages present the risk one takes if the actions is lacking (e.g. Davis 1995; Chang 2007; Maheswaran and Meyers-Levy 1990). However, in the case of a corporation or industry initiating a CSR effort, the negatively framed message would present how an initiative would lead to a removal of potential harm, as opposed to the consequences of not taking action. Consequently, the designed messages are as presented in table 4.1.

(28)

Table 4.1 Message Framing Design

Positive Message Framing Negative Message Framing

More renewable energy in self-operated offshore fields and onshore plants in Norway.

Through our measures, we will help to increase the proportion of solar and wind power in electricity production.

By 2040, we will increase the use of renewable energy from 20% to 50%.

Less non-renewable energy in self-operated offshore fields and onshore plants in Norway.

Through our measures, we will help reduce the proportion of fossil fuels in electricity production.

By 2040, we will reduce the use of non- renewable energy from 50% to 20%.

To manipulate message source there are two options; to use fictitious corporations or to utilize actual industry actors. For this thesis the latter one was chosen, where message source was manipulated through the use of either a corporate logo (Equinor) or that of an external third- party (Olje- og Energiderartementet). This option offers both advantages and limitations. When choosing to apply actual logos, it is evident that biases will follow. However, the oil and gas industry are such an integral part of the Norwegian economic society, and the welfare thereof, using fictional companies were deemed to complicate the study. That is, if the study were to use fictional corporations it would be a challenge to make people understand the distinction between the two. It would demand a thorough explanation on key features of the corporations, explaining how the two differ from each other, and ultimately explaining Olje- og Energidepartementet and Equinor (or any other Norwegian oil and gas producer). Combining the message source and message framing manipulations, the design of experimental survey material is presented in figure 4.1.

(29)

Figure 4.1 Advertisement Design

Positive Message Framing Negative Message Framing

CorporationExternal Third-Party

(30)

3.2.2 Questionnaire design and recruitment of respondents

To assess that the manipulations satisfied the experimental conditions, respondents were requested to answer three questions for each of the four ads. The first question checked that the correct message source was clearly perceived by all respondents (“Who is the sender?”). The two following questions measured the extent to which the messages presented positively (i.e.

ad emphasizes increased use of renewable energy) or negatively (i.e. ad emphasizes the elimination of environmental damage) framed statements. Outcome framing was assessed by two questions measured on seven-point Likert scales with endpoints “Strongly disagree” (1) and “Strongly agree” (7). See table 4.2 for pretest questionnaire.

For pretests’ comparable to this one, literature recommends a default sample size of 30 participants (Perneger, Courvoisies, Hudelson and Gayet-Ageron 2014, 151). Because the manipulations are intended to differ significantly from each other, it was found appropriate that each individual respondent could answer for all four advertisements. As a consequence of the Covid-19 situation, all respondents were recruited via social media and accessed the questionnaire through a direct anonymous link. Accordingly, convenience sampling was used to gather responses from 30 people (Malhotra et al. 2012, 502). Every respondent was presented with all four advertisements to measure if the ads were successfully manipulated in terms of the positive-negative message framing and message source.

Table 4.2 Pretest questionnaire

Variable Question Scale Measure Source

Message source

How is the sender? NA Equinor/Olje- og

energidepartementet

NA

Message framing

To what extent do you agree with the following statements?

The advertisement emphasizes how their measures will provide more sustainable solutions and increased use of renewable energy.

The advertisement emphasizes how their measures will

eliminate environmental damage and reduce fossil fuel use.

1-7 Strongly

disagree/Strongly agree

Perneger et.

al 2014;

Chang et. al 2015

(31)

3.2.3 Results and discussion

The manipulation of message source was found successful, where 100 percent of respondents reported the message source to be as the manipulation intended (i.e. if the ad was communicated by “Equinor”, all respondents reported Equinor to be the sender). As all data relate to the same group of participants and the objective is to test for differences in the means of the pairs, paired samples t-test was used (Malhotra et al. 2012, 647) to assess the manipulation effectiveness of message framing.

Table 4.3 presents the results for the positively framed advertisements. These results indicate that respondents perceived the advertisement in the positive/corporation (M=5.83, SD=1.15) condition to be significantly more positively framed than the negative/corporation (M=3.47, SD=1.74) condition; t (29) = 6.30, p =.000. The same result was found for the third-party source, where the positive/third-party (M=5.67, SD=1.27) condition resulted in significantly higher agreeableness to the statement than the negative/third-party (M=3.70, SD=1.78) condition; t (29) = 5.72, p=.000.

Table 4.3 Manipulation Results for Positive Frame

Paired samples statistics Paired t-test

Positive Negative

Statement (positive frame) Pair M SD M SD t (29) Sig.

Legger vekt på mer bærekraftige løsninger og økt bruk av fornybar energi.

1 Corporation 5.83 1.147 3.47 1.737 6.295 .000 2 Third-party 5.67 1.269 3.70 1.784 5.717 .000 Note: N = 30, M = Mean, SD = Standard deviation. Mean score ranges from 1 (strongly disagree) to 7 (strongly agree), score is calculated from answers given in relation to the relevant statement.

As results in table 4.4 indicate, the negatively framed advertisements were successfully manipulated as well. There was a significant difference in the scores for negative/corporation (M=5.00, SD=1.90) and positive/corporation (M=3.17, SD=1.72) conditions; t (29) = 5.25, p=.000. Furthermore, the respondents perceived the advertisements in the negative/third-party (M=5.47, SD=1.55) condition to be significantly more negatively framed than the positive/third-party (M=3.67, SD=1.87) condition, t (29) = 4.32, p =.000. In conclusion, the

(32)

Table 4.4 Manipulation Results for Negative Frame

Paired samples statistics Paired t-test

Negative Positive

Statement (negative frame) Pair M SD M SD t (29) Sig.

Legger vekt på eliminering av miljøskader og redusert bruk av fossile brennstoff.

3 Corporation 5.00 1.894 3.17 1.724 5.248 .000 4 Third-party 5.47 1.548 3.67 1.868 4.323 .000 Note: N = 30, M = Mean, SD = Standard deviation. Mean score ranges from 1 (strongly disagree) to 7

(strongly agree), score is calculated from answers given in relation to the relevant statement.

3.3 Experimental Study Design

3.3.1 Research design

The purpose of the study is to examine the effect of message source and message framing on attitude towards advertisement and towards the message source. Founded in relevant literature, this thesis suggests that third-party external sources and negatively framed messages will result in a more positive effect on attitude than that of a corporation source and positively framed messages. Furthermore, past literature suggests that this relationship is moderated and mediated by, respectively, level of environmental concern and felt authenticity.

To further examine these relationships, the thesis will introduce a two-factor design; 2 (Message source: Corporation vs. Third-Party) x 2 (Message framing: Positive vs. Negative) between subject design. By use of statistical experimental factorial design, the effects of two independent variables at two different levels are measured, allowing for interactions between variables (Malhotra et al. 2012, 392). Table 4.5 presents the four experimental groups.

Table 4.5 Experimental Groups Message framing

Positive frame Negative frame

Message source

Corporation Experiment group 1 Experiment group 2

Third-party Experiment group 3 Experiment group 4

(33)

3.3.2 Data Collection and Measures

The research questionnaire was created by use of the Qualtrics website (see full questionnaire in appendix B). As mentioned previously the Covid-19 situation posed problems related to recruitment, leading me to recruit respondents by distributing a direct anonyms link via social media. Consequently, responses were gathered through convenience sampling (Malhotra et al.

2012, 502).

As no questionnaire should be distributed and used in experiments without advocate pilot- testing (Malhotra et al. 2012, 477) my thesis advisor, as well as classmates, friends, and family helped identify and eliminate potential problems. To assess the quality of the first questionnaire design, a draft (including demographics) was distributed to my thesis advisor for a review and feedback. After adjustments, a direct link to the questionnaire was sent to a small sample of participants able and willing to point out possible ambiguities and shortcomings for the purpose of improvement. The full questionnaire design can be found in appendix B.

Following the direct-link, respondents were initially presented with the cover story describing the purpose of the questionnaire, including information regarding the expected time duration of questionnaire, and assurance of anonymity. Next, respondents were asked to answer on five 7- point semantic differential scales (bipolar labels) measuring the proposed moderating variable level of environmental concern. Together, these scales ultimately measured their involvement and preoccupation in climate and environmental questions and issues. Following, each respondent was randomly presented with one out of the four CSR advertisements and asked to answer questions directly or indirectly related to the advertisement.

To measure attitude towards CSR messages, attitude was operationalized in two ways; as attitude towards the advertisement and attitude towards the source of the message, yielding two dependent variables (DV). For both DVs, respondents were asked to report on their sentiments connected to the presented advertisement. Attitude is influenced by individuals’ evaluation of the behavior of others, evaluating the behavior as favorable or unfavorable (Ajzen and Cote 2008, 301). This is highly reflected in the scales used in previous research papers (Muehling and Laczniak 1998, 27; Bickart and Ruth 2012, 66; Ajzen and Cote 2008, 301; Groza et al.

2011, 650), as well as in this thesis. Both attitudes towards the CSR campaign and the

Can the Norwegian Oil and Gas Industry Benefit From Communicating their Environmental CSR Efforts?

MAS5100

Masteroppgave

CSR IN CONTROVERSIAL INDUSTRIES:

Can the Norwegian Oil and Gas Industry Benefit From Communicating their Environmental CSR Efforts?

Table of Contents

1.0 Introduction

2.0 Framework

3.0 Methodology and research design